Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Developing Data ProductsUber Tech TalkPete Skomoroch @peteskomorochDecember 5 2012©2012 LinkedIn Corporation. All Rights R...
Examples, Techniques, & Lessons LearnedDeveloping Data Products
Our Mission                          Connect the world’s professionals to make them                             more produ...
LinkedIn is the leading professional network site         187M+                                           1           Link...
LinkedIn profiles represent our professional identity                                 1                             2     ...
We have a lot of data.©2012 LinkedIn Corporation. All Rights Reserved.
We have a lot of data.    And (like everyone else), we store it in Hadoop.©2012 LinkedIn Corporation. All Rights Reserved.
We have a lot of data.  And (like everyone else), we store it in Hadoop.  And people build awesome things with that data.©...
What do we mean by dataproducts?
Building products from data at LinkedInA few examples:    People You May Know    Skills and Endorsements    Year in Rev...
Collaborative Filtering: LinkedIn Skill Pages©2012 LinkedIn Corporation. All Rights Reserved.
Classification: giving structure to unstructured data          Extract©2012 LinkedIn Corporation. All Rights Reserved.
Clustering & Disambiguation©2012 LinkedIn Corporation. All Rights Reserved.
De-duplication and Normalization©2012 LinkedIn Corporation. All Rights Reserved.
Network Algorithms: Relevance & Ranking©2012 LinkedIn Corporation. All Rights Reserved.   15
Prediction: Personalized Skill Recommendations©2012 LinkedIn Corporation. All Rights Reserved.
Skill Endorsements©2012 LinkedIn Corporation. All Rights Reserved.
Social Proof and the Skill Endorsement Graph©2012 LinkedIn Corporation. All Rights Reserved.   20
The Economic Graph: Skills, Jobs, People, Locations…                                                   Location©2012 Linke...
Lessons learned developing dataproducts
Collect the right data at the right time
Large amounts of data can reveal new patterns Probability of Job Title                                                   M...
Be wary of “black-box” approaches©2012 LinkedIn Corporation. All Rights Reserved.   25
Look at your data©2012 LinkedIn Corporation. All Rights Reserved.   26
Aggregate statistics can be misleading       12       10        8        6        4        2        0                 1   ...
Build a viewer app, “micro-listen”©2012 LinkedIn Corporation. All Rights Reserved.   28
Algorithmic intuition: include data geeks in design©2012 LinkedIn Corporation. All Rights Reserved.      29
OODA: Think like a jet fighter©2012 LinkedIn Corporation. All Rights Reserved.   30
OODA: Observe, Orient, Decide, Act©2012 LinkedIn Corporation. All Rights Reserved.   31
OODA: The speed you can move determines victory©2012 LinkedIn Corporation. All Rights Reserved.   32
Red teaming: what can go wrong likely will©2012 LinkedIn Corporation. All Rights Reserved.   33
Error data is super valuable, analyze it and adapt©2012 LinkedIn Corporation. All Rights Reserved.     34
Conclusion: tips for developing data products    Collect the right data at the right time    Large amounts of data can r...
Questions?More info: data.linkedin.com@peteskomoroch©2012 LinkedIn Corporation. All Rights Reserved.   36
Developing Data Products
Developing Data Products
Upcoming SlideShare
Loading in …5
×

Developing Data Products

9,614 views

Published on

Examples, techniques, and lessons learned building data products over the last 3 years at LinkedIn.

Pete Skomoroch is a Principal Data Scientist at LinkedIn where he leads a team focused on building data products leveraging LinkedIn's powerful identity and reputation data.

The talk describes some techniques and best practices applied to develop products like LinkedIn Skills & Endorsements.

This was the inaugural UberData Tech Talk, held in SF at Uber HQ.

Developing Data Products

  1. 1. Developing Data ProductsUber Tech TalkPete Skomoroch @peteskomorochDecember 5 2012©2012 LinkedIn Corporation. All Rights Reserved.
  2. 2. Examples, Techniques, & Lessons LearnedDeveloping Data Products
  3. 3. Our Mission Connect the world’s professionals to make them more productive and successful.Our VisionCreate economic opportunity for every professional in the world.Members First!
  4. 4. LinkedIn is the leading professional network site 187M+ 1 LinkedIn Members 2 640M+ Worldwide Professionals 2 3,300M+ Worldwide Workforce©2012 LinkedIn Corporation. All Rights Reserved. 4
  5. 5. LinkedIn profiles represent our professional identity 1 2 187M Members 187M Member Profiles©2012 LinkedIn Corporation. All Rights Reserved. 5
  6. 6. We have a lot of data.©2012 LinkedIn Corporation. All Rights Reserved.
  7. 7. We have a lot of data. And (like everyone else), we store it in Hadoop.©2012 LinkedIn Corporation. All Rights Reserved.
  8. 8. We have a lot of data. And (like everyone else), we store it in Hadoop. And people build awesome things with that data.©2012 LinkedIn Corporation. All Rights Reserved.
  9. 9. What do we mean by dataproducts?
  10. 10. Building products from data at LinkedInA few examples: People You May Know Skills and Endorsements Year in Review Network Updates Digest InMaps Who’s viewed my profile Collaborative Filtering Groups You May Like and more…©2012 LinkedIn Corporation. All Rights Reserved.
  11. 11. Collaborative Filtering: LinkedIn Skill Pages©2012 LinkedIn Corporation. All Rights Reserved.
  12. 12. Classification: giving structure to unstructured data Extract©2012 LinkedIn Corporation. All Rights Reserved.
  13. 13. Clustering & Disambiguation©2012 LinkedIn Corporation. All Rights Reserved.
  14. 14. De-duplication and Normalization©2012 LinkedIn Corporation. All Rights Reserved.
  15. 15. Network Algorithms: Relevance & Ranking©2012 LinkedIn Corporation. All Rights Reserved. 15
  16. 16. Prediction: Personalized Skill Recommendations©2012 LinkedIn Corporation. All Rights Reserved.
  17. 17. Skill Endorsements©2012 LinkedIn Corporation. All Rights Reserved.
  18. 18. Social Proof and the Skill Endorsement Graph©2012 LinkedIn Corporation. All Rights Reserved. 20
  19. 19. The Economic Graph: Skills, Jobs, People, Locations… Location©2012 LinkedIn Corporation. All Rights Reserved. 21
  20. 20. Lessons learned developing dataproducts
  21. 21. Collect the right data at the right time
  22. 22. Large amounts of data can reveal new patterns Probability of Job Title Months since graduation©2012 LinkedIn Corporation. All Rights Reserved. 24
  23. 23. Be wary of “black-box” approaches©2012 LinkedIn Corporation. All Rights Reserved. 25
  24. 24. Look at your data©2012 LinkedIn Corporation. All Rights Reserved. 26
  25. 25. Aggregate statistics can be misleading 12 10 8 6 4 2 0 1 2 3 4 5 6 7 8 9 10©2012 LinkedIn Corporation. All Rights Reserved. 27
  26. 26. Build a viewer app, “micro-listen”©2012 LinkedIn Corporation. All Rights Reserved. 28
  27. 27. Algorithmic intuition: include data geeks in design©2012 LinkedIn Corporation. All Rights Reserved. 29
  28. 28. OODA: Think like a jet fighter©2012 LinkedIn Corporation. All Rights Reserved. 30
  29. 29. OODA: Observe, Orient, Decide, Act©2012 LinkedIn Corporation. All Rights Reserved. 31
  30. 30. OODA: The speed you can move determines victory©2012 LinkedIn Corporation. All Rights Reserved. 32
  31. 31. Red teaming: what can go wrong likely will©2012 LinkedIn Corporation. All Rights Reserved. 33
  32. 32. Error data is super valuable, analyze it and adapt©2012 LinkedIn Corporation. All Rights Reserved. 34
  33. 33. Conclusion: tips for developing data products Collect the right data at the right time Large amounts of data can reveal new patterns Be wary of “black box” approaches Look at your raw data Aggregate statistics can be misleading Build and use viewer apps Include data geeks in design process OODA: Think like a jet fighter Red-teaming: anticipate edge cases Find opportunity in your error data©2012 LinkedIn Corporation. All Rights Reserved.
  34. 34. Questions?More info: data.linkedin.com@peteskomoroch©2012 LinkedIn Corporation. All Rights Reserved. 36

×